Re: [Boost-users] [boost] [chrono] v0.4.1 + thread_clock implementation on Windows

Subject: Re: [Boost-users] [boost] [chrono] v0.4.1 + thread_clock implementation on Windows
From: vicente.botet (vicente.botet_at_[hidden])
Date: 2010-06-20 10:34:22

----- Original Message ----- 
From: "joel falcou" <joel.falcou_at_[hidden]>
To: <boost-users_at_[hidden]>
Sent: Sunday, June 20, 2010 1:58 PM
Subject: Re: [Boost-users] [boost] [chrono] v0.4.1 + thread_clock implementation on Windows

> 
> vicente.botet wrote:
>> Joel, arithmetic and integral are not the same thing for me. I don't see how I can make ratio be a model of a MPL Integral constant as ratio is not integral, value_type can not be defined, there is no next, nor prior. Am I missing something?
>>
>>   
> I'm fully aware of that, I was more suggesting that, yes, mpl::plus and 
> etc works on ratio. 

Yes, this could be great. Are you suggesting to rename ration_plus to mpl::plus and require that the parameters are Rational Constants?

> Let's call that a Rational Constant then ;)

What should be for you the requirements for a Rational Constant?

Expression Type Complexity 
r::tag rational_c_tag Constant time. 
r::value_type A boost::rational<integral_type> type Constant time. 
r::num An integral constant expression Constant time. 
r::den An integral constant expression Constant time. 
r::type Rational Constant Constant time. 
r::value_type const c = r()   Constant time. 

Expression  Semantics
r::tag  r's tag type; n::tag::value is n's conversion rank.
r::num  The value of the numerator integral constant.
r::den  The value of the denominator integral constant.
r::type  is_equivalent<r::type,r>::value == true.
r::value_type const c = r()  c.numerator() * r::den == c.denominator() * r::num.

> The alternative is to just rename ratio_plus to plus.

Boost.Chrono follows the C++0x namming. What would be the advantage of renaming it?

Best,
Vicente